METAbits – April 2002

Published in April 2002

Controlled Exposure of Information Assets Breeds Enterprise Success

Our most recent industry study on enterprise analytics indicates that making the enterprise’s data warehouse available to suppliers and partners breeds higher levels of success.
During the next three years, the move to extended relationship management (XRM) solutions will further drive the outward expansion of information supply chains (ISCs), so that operational,
collaborative, and analytical information can be made available to partners, employees, customers, and suppliers.

Measuring the Value of DW Technology (Not Just Projects)

80% of IT shops implementing data warehouses (DWs) select related tools and hardware before determining how their benefit will be measured. We also often hear that DW architecture should not be
based on business needs, but rather an idealized architecture concept. Focusing on the business requirements of a data warehouse must go beyond a list of needs. High-return DW projects require an
emphasis on business improvement versus delivery of reports and access to information. CEOs are saying a firm “no” to any capital expenditure with no associated value metrics.

Signs of Analytic Maturity

Progressive enterprises are seriously evaluating some version of loop closing, or collaboration, between analytic apps and business processes. This notion of collaborative business
intelligence formalizes the process of capturing and using intelligence and knowledge assets – ultimately improving information’s value. Given the costs of support, data quality, and cultural
barriers, enterprises no longer depend on spontaneous benefits from data warehouses, but rather appreciate the long-range value of operationally integrated, actionable analytics.

Proving Data Warehouse ROI via Control Groups

The typical enterprise has business units (finance, marketing, sales, HR, manufacturing, service, etc.) simultaneously engaged in multiple initiatives to improve business performance. Therefore, it
can be difficult to assign any one initiative a precise ROI figure. We recommend that data warehouses (DWs) include control groups of subjects (products, marketing programs, customers, partners,
suppliers, employees, etc.) that are the object of established key performance indicators (KPIs) but not included and left unaffected by DW analysis. With these control groups, the
DW can be credited with specific business performance gains.

Service and Quality Performance Indicators

While the current economy forces enterprises to fixate on revenue and profitability, our research shows that a singular analytic focus on accounting numbers does little to drive them upward.
Instead, enterprises now realize that derivative indicators related to “service” and “quality” are the real knobs, buttons, and levers that an organization can turn, push, and
slide to affect business performance. Mentioned among the most important key performance indicators (KPIs) are service (72%), quality (65%), growth (63%), and profitability (59%).

Analytic Quandary: Consistency vs. Creativity

Enterprises must begin making conscious decisions about their “analytic culture” – that is, where on the spectrum of analytic flexibility/creativity versus control/consistency they want to be.
Spreadsheet and independent data-mart-oriented environments offer the greatest latitude, but result in a lack of reporting consistency (e.g., “Whose numbers are right?”) while
enterprise data warehouse architectures with predefined analytics (e.g., queries, reports, dashboards, alerts) can yield greater collaboration at the expense of analytic experimentation.

When Data Warehouses Abound

Recent research reveals that average Global 2000 enterprises have implemented 4+ distinct data warehouse environments. Most of these are either functionally or geographically diverse, as a result
of inabilities to coordinate efforts, resolve semantic differences, and/or leverage iterative development methods. We are observing a 100% increase this year in enterprises seeking ways to
integrate/consolidate these DW efforts/architectures
(largely driven by cost-cutting mandates), and therefore see a market opportunity for vendors and consultants to develop technologies
and techniques to help enterprises rectify this problem.

Scaling the Analytic Maturity Model

Enterprises vary greatly in how well they leverage their information assets. Those at the top of our Analytic Maturity Model scale excel at enterprise reporting against enterprise analytic
(e.g., operational data store and/or centralized multi-tier data warehouse), applying deep analytics (e.g., data mining) to expose/exploit hidden business relationships and
avert fraud/risk, and implementing dashboards to continually monitor business process performance. Integrating analytic output into operational/strategic processes is what really demonstrates
significant analytic maturity.

Analytic “Optimization”

With the economy sputtering, “optimization” has become the key buzzword in analytic circles. Doing more with less requires analytic horsepower and functionality quite distinct from traditional
query/reporting (a.k.a. “business intelligence”) software. In 2002, we expect to see the highest analytic solution adoption rates in data/text mining and process optimization/automation solutions
targeted at marketing, inventory management, manufacturing, HR management, and partner/supplier relationships. Analytics will become increasingly targeted at processes, not people.

Data Warehouse Success: Hard Facts About Soft Issues

Hard details regarding data warehouse (DW) technology and architecture (e.g., performance, functionality) are often furiously debated by internal factions. Our latest DW best-practices study
indicates enterprises must also heed non-technical “soft” issues to realize high project success levels. We find enterprise DW initiatives must be business-driven and business-aligned, rather
than driven by broad infrastructure objectives (e.g., consolidation, data access). Highly successful DWs show 2x greater end-user involvement throughout the development life cycle
and are predominantly staffed by experienced external practitioners. In addition, higher DW ROI is achieved by enterprises that expose information beyond employees and customers (e.g., partners,

The Case for Centralizing DW Project Management

Our research shows 80% of enterprises have dedicated data warehouse (DW) staff (not just a DBA or business analyst with a hobby). However, only about 20% of these (approx. 16% overall) have an
enterprise (i.e., larger than departmental) charter. We believe enterprises should coordinate analytic initiatives centrally to maximize data integration (for higher-order
analytics) and broadly leverage DW/analytic technologies, techniques, and talent. Centrally managed efforts, while politically challenging, result in improved metadata commonality, business
semantic agreement, and resultant operational coordination.

Getting Smart About Business Intelligence

Our research indicates that users are beginning to suffer more from the limitations of standalone traditional “business intelligence” (i.e., query and reporting) tools. Through 2002/03, we
believe many enterprises will realize greater and more sustainable ROI from enterprise-wide analytic solutions that encompass sophisticated data integration capabilities (e.g., ETL),
cross-functional analytics, a dashboard/scorecard interface, and advanced analytics (e.g., data mining). By 2004/05, enterprises will expect such solutions to generate analytic output that
can be readily integrated into core business processes

Focus on Familiar Sources Breeds Early DW Wins

Recent research indicates that highly successful data warehouse (DW) solutions have concentrated on limiting the number of data sources used. Initially, very successful solutions
concentrate on the use of custom applications as the primary data source, and then in latter phases expand the range of data sources to include commercial applications. We recommend that
organizations achieve early “wins” by sticking to sourcing data they know best (particularly data that has been audited) rather than data from complex, proprietary sources.

Extending Information Beyond Just Customers Breeds Enterprise Success

The ability to form tight relationships by extending information access to all business allies (not just customers) is a bellwether of highly successful enterprises. Our recent in-depth study of
data warehouse (DW) projects uncovered a strong positive correlation between the level of DW success and external data sharing to customers, suppliers, partners, and industry organizations. While
nearly all DW success profiles report sharing data with customers, successful solutions also share a significant proportion of their data with partners, industry organizations, and suppliers.

Sub-Optimal Decision Making Equals Real Life

We estimate that 90% of all business decisions are sub-optimal due to data quality issues. While not always adversely affecting the business in , decisions themselves (and their timeliness) are
regularly affected by information’s availability, accessibility, integrate-ability, completeness, correctness, precision, scale, and depth (history).

Used by permission of Doug Laney, META Group, Inc.
Copyright 2002 © META Group, Inc.


submit to reddit

About Doug Laney

Doug Laney, Vice President and Service Director of Enterprise Analytics Strategies for META Group is an experienced practitioner and authority on business performance management solutions, information supply chain architecture, decision support system project methodology, consulting practice management, and data warehouse development tools. Prior to joining META Group in February 1999, he held positions with Prism Solutions as a consulting practice director for its Central US and Asia Pacific regions, as a methodology product manager, and as a consultant to clients in Latin America. With data warehouse solution involvement in dozens of projects, his field experience spans most industries. Mr. Laney's career began at Andersen Consulting, where he advanced to managing batch technical architecture design/development projects for multimillion-dollar engagements. He also spent several years in the artificial intelligence field, leading the development of complex knowledgebase and natural language query applications. Mr. Laney holds a B.S. from the University of Illinois.